The aim of this project is to investigate the feasibility of making quantitative extrapolations of carcinogenesis data from species to species. We will fit a number of carcinogenesis models to data in cases where data are available for two or more animal species, or one animal species and man. The models make predictions of both dose-response relations and time to tumor relations. The fitted models will then be extrapolated to other species by two methods: the units rule and the bodyweight law. The data and predictions will be compared as to specific values and general agreement in form. An evaluation of situations in which species to species extrapolation is and is not feasible will be made.